I don't get it... how do you have a competition with no data on the competition??


I thought you Krembil folks were familiar with the world of BOINC? Have you not ever looked at the competitions from years past if not understanding how to run a competition?

I was kind of excited to see this happening... but it's hard to be excited when there's nothing to see...

With this low production of new work, it's obvious that the "Computation Moonshot" will be won by the team that best can implement the boinccmd.exe "update" command in a script/batch file, or Windows Task Scheduler

Hello all,

I wanted to take this opportunity to address some of the concerns that have been brought up in this thread. Firstly, our tech team has assured us that there is enough work for the time being, and more work units are set to be released. As is mentioned in the article, this competition is hosted by The Science Commons Initiative. They are tracking progress and are providing the prizes for this competition. We have the pleasure of being the platform they are using to promote students' learning about data science. We have asked them for information on registered teams so that we can track progress and pass on any updates to you all.

Thank you all for your support, and we hope that you can see the potential this competition has to support student learning!

What???? It's clear that there isn't enough work, unless you use the boinccmd "update" feature in a script, or batch file, or Task Scheduler. BOINC itself, as default, extends the time between requests, if no work is delivered, more and more, and it can result in many hours between request attempts. That BOINC "feature" is what many users here, are suffering from now. So, excuse me for being blunt, but the tech team doesn't know what they are talking about, when they say that "there is enough work for the time being."

And, on the Computation Moonshot News page https://www.worldcommunitygrid.org/about_us/article.s?articleId=810 It says the following about work: "Current status: MCM - ready; OPN - ready; SCC - ready; ARP - in the process." The fact is that MCM1 have very little work available, same goes for OPNG, and saying that SCC is "ready", is absolutely not true at all. No SCC tasks has been sent out at all, since no SCC task has been sent back to WCG, since 10/28/2023 either. This Global statistics page, shows the number of tasks for each project, and when they were received back to WCG https://www.worldcommunitygrid.org/stat/viewGlobal.do

Just go back on the pages of each project, and you can easily see that MCM and OPNG for example, only have a fraction of normal number of results returned. When MCM runs perfectly, the number of returned results/day, is over a million/day, often way more than a million/day. Yesterday it was only 409,336. And the number of OPNG tasks yesterday, was only 8,434 returned, when a normal flow would be much, much, higher.(looking from the date when OPN1 stopped being sent out, and only OPNG was being sent)

Sorry, but this is clearly a lie.

Unless you have some secret sharing agreements with a privileged few in place, at least as far as OPNG WUs are concerned, they are still as rare as hen's teeth. Among my 7 GPU capable hosts, I got since Monday two (2) OPNG WUs, one each on a different host, both returned within 20min of being send out, and one of those still "pending validation" ever since.

SCC WUs are likewise non-existent, yet another false promise from WCGs side.

The only WUs that seem to be plentiful at this point are MCM1 for me, but bottom line is that with your promise to provide all those WUs, you seem to have once again bitten off more than you can chew. And likely keep censoring people that remind you of this.

I think (probably along with a lot of other, in particular regulars) that you really need to start to provide more honest assessments and information instead of slapping us in the face all the time...

Ralf

PS: Just edited part of my post as it seems you finally removed the moderation